CS390C: Principles of Concurrency

and Parallelism

Suresh Jagannathan
suresh@cs.purdue.edu
http://www.cs.purdue.edu/homes/suresh

and

Ananth Grama
ayg@cs.purdue.edu
http://www.cs.purdue.edu/homes/ayg

http://www.cs.purdue.edu/homes/suresh/CS390C

CS390C: Principles of Concurrency and Parallelism

 Course Layout
● Introduction to Concurrency and Parallelism
− Motivation
− Platforms
● Basic Concepts
− Interaction Models for Concurrent Tasks
− Elements of Concurrency
● Threads, Co-routines, Events
− Concurrency Control
● Serializability, Atomicity
− Performance Measures

CS390C: Principles of Concurrency and Parallelism

 Course Layout
● System Support for Concurrency and Parallelism
− Scheduling Techniques
− Synchronization Mechanisms (locks, transactions, wait-
notify)
● Data Structures
− Queues, Heaps, Trees
● Algorithms
− Sorting, Graph Algorithms
● Abstractions
− Dataflow, Selective Communication, Continuations
● Formal Methods and Analyses
CS390C: Principles of Concurrency and Parallelism
 Grading and Evaluation
● Three to four small programming projects
● One larger project, designed and implemented
in consultation with the instructor(s)
● One final exam

CS390C: Principles of Concurrency and Parallelism

 Introduction to Concurrency and
Parallelism
● What is Concurrency?

Traditionally, the expression of a task in the form

of multiple, possibly interacting subtasks, that
may potentially be executed at the same time
constitutes concurrency.

CS390C: Principles of Concurrency and Parallelism

 Introduction to Concurrency and
Parallelism
● What is Concurrency?

− Concurrency is a programming concept.

− It says nothing about how the subtasks are actually
executed.
− Concurrent tasks may be executed serially or in
parallel.

CS390C: Principles of Concurrency and Parallelism

 Why Concurrency?
● Concurrency plays a critical role in serial as well
as parallel/distributed computing environments.

CS390C: Principles of Concurrency and Parallelism

 Why Concurrency?
● In a serial environment, consider the following
simple example of a server, serving requests
from clients (e.g., a web server and web clients)

Non-concurrent
serial server
request 2 request 1
t=0

CS390C: Principles of Concurrency and Parallelism

Let us process requests serially

request 2 request 1
t=0

request 2 request 1
t=6

request 2 request 1
t=8
Total completion time = 8 units, Average service time = (6 + 8)/2 = 7 units
CS390C: Principles of Concurrency and Parallelism
Try a concurrent server now!
request 1

request 2
t=0

request 1

request 2
t=1

request 1

request 2
t=2

CS390C: Principles of Concurrency and Parallelism

We reduced mean service time!

t=3

t=4

t=8
Total completion time = 8 units, Average service time = (4 + 8)/2 = 6 units
CS390C: Principles of Concurrency and Parallelism
 Why Concurrency?
● The lesson from the example is quite simple:
− Not knowing anything about execution times, we
can reduce average service time for requests by
processing them concurrently!
● But what if I knew the service time for each
request?
− Would “shortest job first” not minimize average
service time anyway?
− Aha! But what about the poor guy standing at the
back never getting any service (starvation/
fairness)?
CS390C: Principles of Concurrency and Parallelism
 Why Concurrency?
● Notions of service time, starvation, and fairness
motivate the use of concurrency in virtually all
aspects of computing:
− Operating systems are multitasking
− Web/database services handle multiple concurrent
requests
− Browsers are concurrent
− Virtually all user interfaces are concurrent

CS390C: Principles of Concurrency and Parallelism

 Why Concurrency?
● In a parallel context, the motivations for
concurrency are more obvious:
− Concurrency + parallel execution = performance

CS390C: Principles of Concurrency and Parallelism

 What is Parallelism?
● Traditionally, the execution of concurrent tasks
on platforms capable of executing more than
one task at a time is referred to as “parallelism”
● Parallelism integrates elements of execution --
and associated overheads
● For this reason, we typically examine the
correctness of concurrent programs and
performance of parallel programs.

CS390C: Principles of Concurrency and Parallelism

 Why Parallelism?
● We can broadly view the resources of a
computer to include the processor, the data-
path, the memory subsystem, the disk, and the
network.
● Contrary to popular belief, each of these
resources represents a major bottleneck.
● Parallelism alleviates all of these bottlenecks.

CS390C: Principles of Concurrency and Parallelism

 Why Parallelism?
● Starting from the least obvious:
− I/O (disks) represent major bottlenecks in terms of their
bandwidth and latency
− Parallelism enables us to extract data from multiple
disks at the same time, effectively scaling the
throughput of the I/O subsystem
− An excellent example is the large server farms (several
thousand computers) that ISPs maintain for serving
content (html, movies, mail).
− Try logging into mail.yahoo.com and see where your
mail is hosted at (mine is at us.f301.mail.yahoo.com)
CS390C: Principles of Concurrency and Parallelism
 Why Parallelism?
● Most programs are memory bound – i.e., they operate at a
small fraction of peak CPU performance (10 – 20%)
● They are, for the most part, waiting for data to come from
the memory.
● Parallelism provides multiple datapath's to memory –
effectively scaling memory throughput as well!

CS390C: Principles of Concurrency and Parallelism

 Why Parallelism?
● The process itself is the most obvious bottleneck.
● Moore's law states that the component count on a die
doubles every 18 months.
● Contrary to popular belief, Moore's law says nothing about
processor speed.
● What does one do with all of the available “components”
on the die?

CS390C: Principles of Concurrency and Parallelism

 Parallelism in Processors
● Processors increasingly pack multiple cores into a single
die.
Why?

CS390C: Principles of Concurrency and Parallelism

 Parallelism in Processors
● The primary motivation for multicore processors, contrary
to belief is not speed, it is power.
● Power consumption scales quadratically in supply voltage.
● Reduce voltage, simplify cores, and have more of them –
this is the philosophy of multicore processors

CS390C: Principles of Concurrency and Parallelism

 Why Parallel?
● Sometimes, we just do not have a choice – the data
associated with the computations is distributed, and it is
not feasible to collect it all.
− What are common buying patterns at Walmart across
the country?
● In such scenarios, we must perform computations in a
distributed environment.

CS390C: Principles of Concurrency and Parallelism